✨Data Intelligence Lab@HKU✨

A tour of the ingestion pipeline, multimodal processors, and resilience machinery that turn messy files into graph-ready knowledge.

A deep dive into the agentic pipeline, the camera tree, and the structured shot logic that try to solve the hardest problem in video generation: staying coherent from scene to scene.

HKUDS splits planning, execution, and review into separate roles, then uses a shared board to keep the whole system legible, recoverable, and easier to debug.

A self-hosted agent runtime with multi-channel reach, MCP-native tools, and a Dream memory system that turns long-term context into readable files instead of opaque state.

An open-source messenger built for AI agents, human teams, and the messy middle between them.

HKUDS turns trading into a social protocol. Agents read `SKILL.md`, register themselves, publish signals, debate strategy, and execute across stocks, crypto, and Polymarket.

Instead of rethinking every screen with a vision model, PhoneCLI maps mobile navigation once, replays it as deterministic commands, and falls back to a model only when the UI is genuinely unfamiliar.

It records desktop activity as events, closes them into hierarchical summaries, and gives agents a way to browse your day without a cloud vector index.

HKUDS's open-source system turns repositories into layered structure, then retrieves just enough context to answer a question without flooding the model.

HKUDS built a multi-agent finance workspace that turns natural language into backtests, market-specific code, and cross-platform exports for global trading workflows.

HKUDS built a universal tool-use layer that narrows a noisy tool universe to the few calls that actually matter, then routes work through GUI, shell, MCP, or a sandboxed server.

A deep dive into the control plane behind tool use, memory, permissions, and multi-agent coordination.

The open-source editor pairs real-time collaboration, AI completion, and a shadow-document workflow so researchers can draft without breaking the source.

HKUDS’s evaluation framework scores long AI research reports for structure, insight, redundancy, and factual grounding by pairing LLM judges with live web verification.

HKUDS turns agent evaluation into a simulated economy, with task bounties, inference costs, occupation-specific judging, and a survival rule that punishes waste.

How HKUDS is solving the LLM contamination crisis by forcing models to bet against the real-world wisdom of the crowd.

How HKUDS is stripping the visual skin off professional software to arm the next generation of AI agents.

How HKUDS and Alibaba are moving beyond "Deep Research" to automate the "Aha!" moment of scientific discovery through decoupled reinforcement learning.

An open-source tutoring system that can spin up specialist agents, preserve a clean student transcript, and build a lasting memory of what each learner actually knows.

How a tiny model's failures teach a giant model to reason with 99% less data.

A deep dive into a self-evolving middleware for Claude Code, Cursor, and other MCP agents, where skills are repaired, reused, and ranked like living code.

How a research lab bypassed complex API orchestration by using tmux, git worktrees, and filesystem locks to manage autonomous AI agents.